When DNA is damaged, cells initiate a signaling cascade that results in cell cycle checkpoint arrest and repair of the damage or elimination of the damaged cells. One pathway that is activated following a double strand DNA break involves ATM and other DNA damage response proteins including CHEK2, Cdc25A, Plk3, BRCA1 and BRCA2. Several genes in this pathway, which contribute to genomic stability, have variant alleles that predispose to disease. This study focuses on mouse models and human cohorts who have had prior radiation exposure. The hypothesis is that variant alleles within the DNA damage response pathway may be benign or produce a modest risk, but that their adverse effects manifest more profoundly after exposure. One mouse model mimicks a CHEK2 variant that predisposes to breast cancer. Mice homozygous and heterozygous for this allele will be assessed for their inherent DNA repair capacity and whether they are more susceptible to disease and to mutagenesis, particularly to mitotic recombination and LOH before and after exposure. Other mouse models with deficiencies or variants in Plk3, BRCA2, and Cdc25A, have been made or are under construction. Additional models, dictated by the outcome of the human studies, will be produced. The human studies will utilize two unique cohorts with prior environmental exposure in the form of radiation who subsequently developed breast cancer. A control population consists of individuals who have experienced similar exposure but have not developed disease. The first cohort consists of patients who as children were treated with radiation for Hodgkin disease and years later developed breast cancer. The second cohort involves patients who worked at or lived in close proximity to the Fernald Uranium processing plant and have developed breast cancer. Specifically, we will ask whether variant alleles within the DNA damage response pathway, singly or in combination, are overrepresented in the affected population. When such alleles are identified, they will be modeled and tested in the mouse. Lastly, cells from affected individuals will be tested for genomic instability. In brief, this project assesses allelic variants within a pathway in two unique exposed populations and models the genotypes in mice. Since this pathway responds to damage induced by oxidative stress and since these alleles may sensitize to oxidative stress, the capacity of antioxidants to prevent disease in mouse models carrying variant alleles will be tested.